cell phone use and behavioral problems in children: an...
Post on 30-May-2020
6 Views
Preview:
TRANSCRIPT
Review Copy
Cell Phone Use and Behavioral Problems in Children: An
Analysis Based on Prospective Cohort
Journal: Pediatrics
Manuscript ID 2016-0107
Article Type: Regular Article
Date Submitted by the Author: 12-Jan-2016
Complete List of Authors: Sudan, Madhuri; University of California, Los Angeles, Department of Epidemiology Olsen, Jørn; University of Aarhus, Institute of Public Health, Department of Epidemiology Arah, Onyebuchi; UCLA Fielding School of Public Health, Department of Epidemiology
Obel, Carsten; University of Aarhus, Institute of Public Health, Department of Epidemiology Kheifets, Leeka; UCLA School of Public Health, Epidemiology;
Keyword/Topic: Psychosocial Issues < Developmental/Behavioral Issues, Environmental Health
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
Review Copy
1
Cell Phone Use and Behavioral Problems in Children: An Analysis Based on Prospective
Cohort
Madhuri Sudan
a,b, Jorn Olsen
b, Oyebuchi A. Arah
a, Carsten Obel
c, and Leeka Kheifets
a*
Affiliations: a Department of Epidemiology, Fielding School of Public Health, University of
California, Los Angeles (UCLA), Los Angeles, CA 90024, USA; b Danish Epidemiology
Science Centre, Department of Public Health, Aarhus University, 8000 Aarhus C, Denmark; c
Institute of General Medical Practice, Center for Collaborative Health, Aarhus University, 8000
Aarhus C, Denmark
Address correspondence to: Leeka Kheifets, 650 Charles E Young Drive S, 71-279 CHS,
Los Angeles, CA 90095, 310-825-6950, kheifets@ucla.edu
Short title: Cell Phone Use and Behavioral Problems in Children
Financial Disclosure: The authors have no financial relationships relevant to this article to
disclose.
Funding source: This work was supported by funding from the European Community’s Seventh
Framework Programme FP7/2007-2013 [603794 – the GERONIMO project].
Potential Conflicts of Interest: The authors have no conflicts of interest relevant to this article
to disclose.”
Abbreviations: Strengths and Difficulties Questionnaire (SDQ); Odds Ratio (OR); Confidence
Interval (CI); radio frequency (RF) fields.
What’s Known on This Subject:
Children are increasingly exposed to cell phones beginning at very early ages, including in utero.
Many parents and pediatricians are concerned that this technology could have negative health
effects, but studies of children’s exposure are few.
What This Study Adds:
This is the first large scale cohort study to prospectively examine cell phone use (during
pregnancy and at age 7) and behavioral problems (at age 11). Our findings support the link
between cell phone use and behavioral problems in children.
Page 1 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
2
Contributors' Statement:
Dr. Kheifets conceptualized and designed the study, edited the initial manuscript, reviewed and
revised the subsequent versions, and approved the final manuscript as submitted.
Dr. Olsen conceptualized and designed the study, coordinated and supervised data collection,
reviewed and revised the manuscript, and approved the final manuscript as submitted.
Dr. Sudan carried out the initial analyses, drafted and revised the manuscript, and approved the
final manuscript as submitted.
Dr. Arah participated the analyses, reviewed and revised the manuscript, and approved the final
manuscript as submitted.
Dr. Obel designed the data collection instruments, critically reviewed the manuscript, and
approved the final manuscript as submitted.
All authors approved the final manuscript as submitted and agree to be accountable for all
aspects of the work.
Page 2 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
3
Abstract (207)
Background
We previously reported associations between cell phone use and behavioral problems in children
in the Danish National Birth Cohort using cross-sectional data. To overcome the limitations of
cross-sectional analysis, we examine these associations with prospectively collected data.
Methods
Based on maternal reports, prenatal and postnatal cell phone use was assessed at 7 years, and
behavioral problems were assessed at 7 and 11 years with the Strengths and Difficulties
Questionnaire. We performed logistic regression analyses to estimate odds ratios and 95%
confidence intervals relating prenatal and age-seven cell phone use to behavioral problems at age
11.
Results
Among children at risk without behavioral problems at age seven), those with both prenatal and
age-seven cell phone use had the highest odds of behavioral problems at age 11 (OR: 1.58; 95%
CI: 1.34–1.86), followed by those with prenatal use only (OR: 1.41; 95% CI: 1.20–1.66) and
age-seven use only (OR: 1.36; 95% CI: 1.14–1.63). These results did not materially change when
early adopters were excluded or when children with behavioral problems at age seven were
included in the analysis.
Conclusions
Our findings are consistent with patterns seen in our earlier studies and suggest that both prenatal
and postnatal use are associated with behavioral problems in children.
Keywords: children, cell phones, behavior
Page 3 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
4
Introduction
Cell phone use has increased rapidly in recent years. Between 2000 and 2014, the number of
active cell phone subscriptions increased from 700 million to nearly seven billion globally (1, 2),
and cell phones are the highest localized source of exposure to radiofrequency fields (RF) (3, 4).
While the health effects of this exposure remain uncertain, there is concern about a possible
effect on vulnerable populations such as children. The World Health Organization (5) and the
National Academy of Sciences (6) have identified prospective cohort studies of RF exposure and
neurological outcomes among children as a high priority research need. Children may be at
increased risk of potential health effects due to developing organ and tissue systems, particularly
the nervous system, and children have higher specific absorption rates of RF than adults (7, 8).
Children born in the last 20 years are part of a unique population having been exposed to cell
phones starting in early life, and they will probably continue to be exposed throughout their lives
and experience a much higher lifetime exposure than seen before.
Studies among children and adolescents, including work by our group, have found associations
between cell phone use and changes in behavior and cognitive function (9-12). In our first study
of 13 159 children in the Danish National Birth Cohort (DNBC), we found associations between
prenatal cell phones use and behavioral problems at age 7 years (12). The strongest association
was seen among children who were exposed both prenatally and postnatally, with an odds ratio
(OR) of 1.80 and a corresponding 95% confidence interval (CI) of 1.45 to 2.23 when compared
to those with no use. The results also suggested a small positive association of postnatal-only use
with behavioral problems. To address the issue of possible confounding, we replicated these
Page 4 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
5
results in a separate group of 7-year-old children in the DNBC (n = 28 745). While the
associations were weaker, they remained after controlling for several additional confounding
factors (10). After excluding twins and triplets, the two datasets were then combined and the
results replicated in a group of 41 541 children, revealing consistent associations with this
enhanced statistical power. Further, these associations were not limited to early adopters of the
technology, and these findings were supported by two smaller studies (9, 11). Investigators of
another small study concluded that they did not find any association between prenatal cell phone
use and behavioral problems in children, but their results were not inconsistent with our findings
(confidence intervals overlapped) (13, 14).
A major limitation of our previous investigations was that behavioral problems were assessed at
the same time as the assessment of cell phone use. With cross-sectional assessment, recall bias or
reverse causation were potential sources of bias. A new wave of data collection was recently
completed in the DNBC, and it included an assessment of behavioral problems in the children at
age 11 years. Using the newly collected outcome data, we have re-examined the associations
between cell phone use and behavioral problems among children with prospective data,
overcoming the key limitation of our previous studies. Specifically, our current investigation
examined associations of prenatal only, postnatal only, and both prenatal and postnatal cell
phone use, assessed at age 7 years, with behavioral problems at age 11 years.
Methods
Study setting and population:
Page 5 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
6
The DNBC enrolled 91 661 pregnant women in Denmark between 1996 and 2002, with 9380
enrolled again during subsequent pregnancies within the enrollment period. Approximately 50%
of all pregnant women in Denmark were invited to participate, and about 60% of those accepted.
A total of 96 841 children were born into the cohort and have been followed since the prenatal
period. For each pregnancy, the DNBC collected detailed information on lifestyle and
environmental exposures by computer-assisted telephone interviews with the women at
gestational weeks 12 and 30, and again post-partum when the children were 6 and 18 months and
7 and 11 years old (15). Information on social conditions, birth outcomes, and hospital diagnoses
recorded in Denmark’s national population registries is also linked to the DNBC.
When the children reached 7 years of age, mothers completed detailed questionnaires focusing
on the child’s health and development. Of the 91 256 mothers invited to participate in the age-
seven wave of data collection, 59 975 completed the self-administered questionnaire (66%
participation rate). This investigation includes 54 908 singleton-born children (e.g., not twins,
triplets, etc.) whose mothers participated in the age-7 wave of data collection. When the children
reached 11 years of age, a new wave of data collection was carried out. Age-11 data collection
was completed in 2014, in which 47 721 mothers participated (52% of those invited).
Exposure:
Information about exposure for this study was drawn from the age-seven DNBC questionnaire.
Mothers responded to questions about the number of times they spoke on their cell phones per
day during her pregnancy as well as whether or not the child used a cell phone more than 1 hour
Page 6 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
7
per week, less than 1 hour per week, or not at all at age 7 years. Children whose mothers reported
using a cell phone during pregnancy were classified as having prenatal use, and children who
reportedly used a cell phone for any amount of time at age 7 years were classified as having
postnatal use. Only a few children used a cell phone for more than 1 hour per week, and thus,
those who used a cell phone for more or less than 1 hour per week were grouped into a single
category for analysis. Prenatal and postnatal (age-seven) cell phone use was grouped into four
categories (0 = cell phones not used, 1 = prenatal use only, 2 = age-seven use only, and 3 = both
prenatal and age-seven use).
Outcomes:
Information was drawn from both the age-seven and age-11 questionnaires, in which behavioral
problems were assessed using the parent version of the 25-question Strengths and Difficulties
Questionnaire (SDQ), a validated standard tool for screening for behavioral problems in children
(16, 17). Mothers responded to 25 statements regarding the child’s behavior on a three-point
scale (1 = not true, 2 = partly true, and 3 = very true). A previously developed algorithm
generated a ‘‘total behavioral difficulties’’ score using responses to 20 of the 25 items in the
SDQ (www.sdqinfo.org). A priori-defined cutoff points for the score were used to classify each
child as ‘‘normal’’ (score = 0-13), ‘‘borderline’’ (score = 14-16), or ‘‘abnormal’’ (score= 17-40)
for overall behavioral problems (16). These cutoff points were based on population-based norms
and are part of the SDQ algorithm, and they are the same cutoff points as those used in our
previous studies (10). ‘Overall behavioral problems at age 11 years’ was the main outcome of
Page 7 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
8
interest in this study. Information about behavioral problems at age 7 years was used to define
subgroups in the analysis.
Statistical analysis:
To test our primary hypothesis, our main analysis focused on the subgroup of children who had
“normal” behavior at age 7 years based on the SDQ (the cohort at risk), and followed this cohort
of children prospectively to age 11 years. By excluding prevalent cases, we examined the change
from a normal to an abnormal score as a function of cell phone use. We used logistic regression
models to compute OR’s and 95% CI’s for the associations of prenatal only, age-seven only, and
both prenatal and age-seven cell phone use assessed at age 7 years with behavioral problems at
age 11 years. We adjusted for the following potential confounders: sex of child, mother’s age at
birth, mother’s and father’s history of psychiatric, cognitive, or behavioral problems as a child,
socio-occupational status, gestational age at birth, mother’s prenatal stress, and breastfeeding.
We chose these variables for adjustment in order to corroborate the results of our second cross-
sectional analysis by Divan et al. (10).
We also examined in more detail the associations between maternal phone use characteristics
during pregnancy and behavioral problems at age 11 years in the cohort at risk. As in the
previous analysis, the specific phone use characteristics examined were the number of times per
day the mother spoke on her cell phone, the percentage of time her phone was powered on when
not in use, and whether or not she used a hands-free device during pregnancy.
Page 8 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
9
We performed a sensitivity analysis by repeating our main analysis in different subgroups. First,
we conducted the analysis in the cohort of children at risk, but excluded those born prior to 1999
(early adopters). Second, we performed the analysis in all children with information on cell
phone use at age 7 years, regardless of their behavioral problems status at age 7 years.
The Danish Data Protection Agency, the regional science ethics committees in Denmark, and the
Office for the Protection of Research Subjects at the University of California, Los Angeles all
approved this study. Women who participated in the DNBC gave written informed consent prior
to inclusion in the cohort. Women who requested to discontinue participation at any time or
whose child was deceased were not contacted for further follow-up.
Results
Among DNBC children who were at risk of behavioral problems at 7 years of age, 28 139 (55%)
used cell phones, with 21% prenatally use only, 16% used cell phones postnatally at age 7 years
only, and 19% used both prenatally and postnatally at age 7 years. Approximately 41% did not
use cell phones during either time period, and 4% had missing information about use in the age-
seven questionnaire (Table 1). Of the children in this cohort at risk whose mothers completed the
subsequent SDQ when the children were 11 years old, 2% scored as having abnormal behavior,
3% as borderline, and 95% as normal.
[Table 1 here]
Page 9 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
10
Among children at risk for behavioral problems at age 7 years, prenatal cell phone use and
postnatal use at age 7 years were both associated with behavioral problems at age 11 years
(Table 2). Children with any postnatal use at age 7 years had 23% higher odds of behavioral
problems than children without postnatal use, while any prenatal use was associated with 30%
higher odds of behavioral problems. The strongest association between exposure and behavioral
problems was seen among children who were exposed both prenatally and at age seven (OR:
1.58; 95% CI: 1.34–1.86), followed by those with prenatal use only (OR: 1.41; 95% CI: 1.20-
1.66), and age-seven use only (OR: 1.36; 95% CI: 1.14-1.63) compared to children with no
exposure.
[Table 2 here]
Specific maternal prenatal cell phone use characteristics were not associated with behavioral
problems in children at age 11 years (Table 3). No clear “exposure-response” trend was seen for
frequency of use, percentage of time the phone was powered on, or use of a hands-free device in
relation to behavioral problems at age 11 years. However, using a hands-free device “often” was
associated with higher odds of behavioral problems.
[Table 3 here]
When we repeated the main analysis in the cohort at risk while excluding early adopters, and
again among all children including those with behavioral problems at age 7 years (Table 4), the
associations did not change much, and the pattern in the results remained the same as in the
Page 10 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
11
analysis within only the cohort at risk.
[Table 4 here]
The results from the prospective analysis were also very similar to the findings from the previous
cross-sectional analysis from Divan et al. (Table 5). OR’s for behavioral problems at age 11
years were very similar to those for behavioral problems at age 7 years among children with
prenatal only use or both prenatal and age-seven use compared to those with no use. However,
children who only used cell phones at age 7 years had slightly higher odds of behavioral
problems at age 11 years (OR: 1.4; 95% CI: 1.1-1.6) than at age 7 years (OR: 1.2; 95% CI: 1.0-
1.3).
[Table 5 here]
Discussion
In this investigation, we found that cell phone use assessed at age 7 years was associated with
behavioral problems in children at age 11 years. The use of prospectively collected data is one of
the key strengths of this study, as it overcomes many limitations of previous cross-sectional
studies. We do not expect our results to be due to reverse causation or recall bias, since we
examined the development of behavioral problems over time in the subgroup of DNBC children
who started with normal behavior at age 7 years, and the use was assessed 4 years prior to the
outcome.
Page 11 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
12
Our results from a cohort of children at risk for behavioral problems at age 7 years were
consistent with those from our previous cross-sectional analyses. The highest odds of behavioral
problems were seen among children with both prenatal and postnatal use, followed by those with
prenatal use only, and then those with postnatal use only. However, children with postnatal only
use had higher odds of behavioral problems at age 11 years (OR: 1.4; 95% CI: 1.1-1.6) than at
age 7 years in the previous dataset (OR: 1.2; 95% CI: 1.0-1.3) or at age 7 years using the current
dataset (OR: 1.02; 95% CI: 0.9-1.2). This could suggest that the influence of postnatal use on
behavioral problems did not manifest until age 11. No exposure-response trends were seen
between maternal prenatal cell phone use behaviors and behavioral problems at age 11 years,
although using a hands-free device “often” increased the risk estimate, consistent with higher
exposure to the fetus.
Our results did not change materially when we excluded early adopters from the cohort at risk
nor when we included all children in our analysis (including those with behavioral problems at
age 7 years). Overall, our findings suggest that both prenatal and postnatal use increase the risk
of behavioral problems in children, but as the length of time since prenatal use increases, the
impact of prenatal use on behavioral problems decreases, while the role of postnatal use
increases.
The DNBC is a large and well-documented birth cohort, and we were able to draw on many
sources of data on potential confounders, including prenatal and early-life interviews and
national Danish social and medical registers. We adjusted for the same potential confounders as
Page 12 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
13
in the previous investigation, thus replicating the previous models using prospective data. For
additional uncontrolled confounding to explain our results there must exist a strong unmeasured
confounder that is closely related to both cell phone use and behavioral problems, and we believe
this is unlikely to be the case.
About 33% of children for whom cell phone use was assessed at age 7 years had missing
information about behavioral problems at age 11, with a slightly larger proportion of those with
both prenatal and postnatal cell phone use missing these data. This was mostly due to loss-to-
follow-up. Previous research in the DNBC found that women who were lost to follow-up in
earlier waves of data collection were more likely to be in the low socio-occupational status
category than those who continued participation (18). Further, some studies reported that social-
occupational status was inversely associated with cell phone use among children and adolescents
(19-22). Therefore, it is possible that children (and possibly mothers) who are heavier cell phone
users may have been less likely to continue follow-up in the DNBC, and therefore are
underrepresented in our data. This could have biased our results, and the magnitude and direction
of this potential bias need to be evaluated further. We do not expect bias due to loss to follow-up
or other sources of missing data to account for the consistent associations we observed in this
investigation.
Conclusion
Our findings give further support to the link between cell phone use and the development of
behavioral problems in children. If our results reflect a casual effect, at least two mechanisms are
possible. One possibility is that RF exposure from cell phones is a possible cause through a yet to
Page 13 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
14
be identified underlying biological mechanism. Another possible explanation is that the use of
cell phones itself (and not RF) leads to altered social development and detrimental changes in
behavior. Research suggests that individuals who are heavy media multi-taskers are more
susceptible to distraction and have difficulty filtering out irrelevant environmental stimuli (23),
and certain cell phone activities have been linked to lower academic performance and higher
levels of anxiety among college students (24). Altered brain function in children and adolescents
in response to heavy cell phone use and use of other media is not surprising as neural plasticity is
very high in these age groups. However, it is not clear to what extent cell phone use results in
normal brain adaptation to new stimuli versus abnormal developmental processes, as this is
likely to depend on the specific context and age of the child (25).
In the coming years, cell phone use will continue to become increasingly prevalent among
children. With the popularity of smartphones, increasing numbers of children will be sending
emails and text messages, while often simultaneously playing games, browsing the internet, and
performing other activities on their devices. Heavy engagement in these activities and possibly
RF exposure from cell phones could be harmful to human health, and may explain the
associations with behavioral problems we observed in this study. For these reasons, much more
research is needed to understand the behavioral and health effects of this technology in children.
We will continue to examine these issues in our data with other analytic methods including bias
analysis (26, 27). New studies among children with prospective data collection should be
conducted to corroborate our findings and to examine behavior, cognition, and cell phone
exposure in more detail, and research should continue into adulthood.
Page 14 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
15
References
1. International Telecommunication Union. Global mobile cellular subscriptions, total and
per 100 inhabitants, 2000-2010. 2011 August 18.
2. International Telecommunication Union. Key ICT indicators for developed and
developing countries and the world (totals and penetration rates). 2015 February 9.
3. Vrijheid M, Deltour I, Krewski D, Sanchez M, Cardis E. The effects of recall errors and
of selection bias in epidemiologic studies of mobile phone use and cancer risk. J Expo Sci
Environ Epidemiol. 2006; 16(4):371-384.
4. Neubauer G, Feychting M, Hamnerius Y, et al. Feasibility of future epidemiological
studies on possible health effects of mobile phone base stations. Bioelectromagnetics.
2007; 28(3):224-230.
5. World Health Organization. WHO Research Agenda for Radiofrequency Fields. Geneva,
Switzerland: 2010 October 22.
6. Barnes FS, Gandhi OP, Hietanen M, et al. Identification of Research Needs Relating to
Potential Biological or Adverse Health Effects of Wireless Communication Devices.
Washington, D.C.: The National Academies Press, 2008.
7. Kheifets L, Repacholi M, Saunders R, van Deventer E. The sensitivity of children to
electromagnetic fields. Pediatrics. 2005; 116(2):e303-313.
8. Wiart J, Hadjem A, Wong MF, Bloch I. Analysis of RF exposure in the head tissues of
children and adults. Phys Med Biol. 2008; 53(13):3681-3695.
9. Abramson MJ, Benke GP, Dimitriadis C, et al. Mobile telephone use is associated with
changes in cognitive function in young adolescents. Bioelectromagnetics. 2009;
30(8):678-686.
10. Divan HA, Kheifets L, Obel C, Olsen J. Cell phone use and behavioural problems in
young children. J Epidemiol Community Health. 2012; 66:524-529.
11. Thomas S, Heinrich S, von Kries R, Radon K. Exposure to radio-frequency
electromagnetic fields and behavioural problems in Bavarian children and adolescents.
Eur J Epidemiol. 2010; 25(2):135-141.
12. Divan HA, Kheifets L, Obel C, Olsen J. Prenatal and postnatal exposure to cell phone use
and behavioral problems in children. Epidemiology. 2008; 19(4):523-529.
13. Guxens M, van Eijsden M, Vermeulen R, et al. Maternal cell phone and cordless phone
use during pregnancy and behaviour problems in 5-year-old children. J Epidemiol
Community Health. 2013; 67(5):432-438.
14. Sudan M, Kheifets L, Arah OA, Olsen J. On the association of cell phone exposure with
childhood behaviour. J Epidemiol Community Health. 2013; 67(11):979.
15. Olsen J, Melbye M, Olsen SF, et al. The Danish National Birth Cohort--its background,
structure and aim. Scand J Public Health. 2001; 29(4):300-307.
16. Goodman R. The Strengths and Difficulties Questionnaire: a research note. J Child
Psychol Psychiatry. 1997; 38(5):581-586.
17. Goodman R, Renfrew D, Mullick M. Predicting type of psychiatric disorder from
Strengths and Difficulties Questionnaire (SDQ) scores in child mental health clinics in
London and Dhaka. Eur Child Adolesc Psychiatry. 2000; 9(2):129-134.
Page 15 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
16
18. Greene N, Greenland S, Olsen J, Nohr EA. Estimating bias from loss to follow-up in the
Danish National Birth Cohort. Epidemiology. 2011; 22(6):815-822.
19. Mezei G, Benyi M, Muller A. Mobile phone ownership and use among school children in
three Hungarian cities. Bioelectromagnetics. 2007; 28(4):309-315.
20. Koivusilta LK, Lintonen TP, Rimpela AH. Orientations in adolescent use of information
and communication technology: a digital divide by sociodemographic background,
educational career, and health. Scand J Public Health. 2007; 35(1):95-103.
21. Bohler E, Schuz J. Cellular telephone use among primary school children in Germany.
Eur J Epidemiol. 2004; 19(11):1043-1050.
22. Thomas S, Heinrich S, Kuhnlein A, Radon K. The association between socioeconomic
status and exposure to mobile telecommunication networks in children and adolescents.
Bioelectromagnetics. 2010; 31(1):20-27.
23. Ophir E, Nass C, Wagner AD. Cognitive control in media multitaskers. P Natl Acad Sci
USA. 2009; 106(37):15583-15587.
24. Lepp A, Barkley JE, Karpinski AC. The relationship between cell phone use, academic
performance, anxiety, and Satisfaction with Life in college students. Computers in
Human Behavior. 2014; 31:343-350.
25. Courage ML, Bakhtiar A, Fitzpatrick C, Kenny S, Brandeau K. Growing up multitasking:
The costs and benefits for cognitive development. Developmental Review. 2015; 35(0):5-
41.
26. Arah OA, Sudan M, Olsen J, Kheifets L. Marginal structural models, doubly robust
estimation, and bias analysis in perinatal and paediatric epidemiology. Paediatric and
perinatal epidemiology. 2013; 27(3):263-265.
27. Arah OA, Chiba Y, Greenland S. Bias formulas for external adjustment and sensitivity
analysis of unmeasured confounders. Ann Epidemiol. 2008; 18(8):637-646.
Page 16 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
17
Table 1. Distributions of covariates and behavioral problems at age 11 years among children at risk of
behavioral problems at age 7 years stratified by cell phone exposure categories (N=51 190)
No exposure
(n = 20 801)
Prenatal
exposure
only
(n = 10 619)
Age-seven
exposure
only
(n = 7942)
Both prenatal
and age-seven
exposure
(n = 9578)
Unknown
exposure
(n = 2250)
Sex
Male 52.4 53.2 45.6 45.6 52.8
Female 47.6 46.8 54.4 54.4 47.2
Age of mother in years
24 or younger 3.3 6.7 4.9 11.3 6.4
25-29 33.1 35.8 35.9 38.9 33.8
30-34 43.2 39.8 40.9 34.4 39.9
35-39 17.8 15.4 15.9 13.5 17.0
40 or older 2.5 2.3 2.3 2.0 2.8
Mother’s history of psychiatric problems
Yes 12.0 13.3 14.7 16.4 13.3
No 88.0 86.7 85.3 83.6 86.7
Socio-occupational levels
High 73.1 72.1 69.9 64.8 66.5
Med 24.7 25.1 27.2 30.5 30.1
Low 2.2 2.8 2.9 4.7 3.4
Gestational age at birth in weeks
<37 weeks 3.1 3.4 3.2 4.0 3.2
37-41 weeks 81.0 80.6 80.7 80.4 81.4
42 or greater 15.9 16.0 16.1 15.6 15.3
Mother’s prenatal stress score
Low (0-4) 93.6 93.1 91.7 90.4 91.0
Medium (5) 3.0 3.6 3.6 4.1 4.4
High (6-14) 3.4 3.3 4.7 5.6 4.6
Mother’s history of psychiatric, cognitive,
or behavioral problems as a child
Yes 12.3 12.2 13.2 14.9 14.7
No 87.7 87.8 86.8 85.1 85.3
Father’s history of psychiatric, cognitive,
or behavioral problems as a child
Yes 9.5 9.3 10.5 11.0 10.1
No 90.5 90.7 89.5 89.0 89.9
Child breastfed up to 6 months of age
Yes 70.4 63.6 67.4 59.7 65.6
No 29.7 36.4 32.6 40.3 34.4
Behavioral Problems at age 11
Normal 96.4 95.0 95.0 94.0 93.6
Borderline 2.1 2.9 2.8 3.6 3.9
Abnormal 1.5 2.1 2.2 2.4 2.5 Results are reported as percentages; Missing values not shown.
Page 17 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
18
Table 2. Associations between prenatal and age-seven cell phone exposure and
overall behavioral problems at age 11 years in children at risk of behavioral
problems at age 7 years
Unadjusted OR (95% CI) Adjusted OR (95% CI)b
Prenatal exposure
No prenatal exposurea 1.00 1.00
Prenatal exposurec 1.32 (1.19–1.47) 1.30 (1.15–1.48)
Postnatal exposure
No age-seven exposurea 1.00 1.00
Age-seven exposured 1.29 (1.16–1.44) 1.23 (1.09–1.40)
Prenatal and/or postnatal exposure
No exposurea 1.00 1.00
Prenatal only 1.40 (1.22–1.61) 1.41 (1.20–1.66)
Age-seven only 1.39 (1.19–1.61) 1.36 (1.14–1.63)
Both prenatal and age-seven 1.69 (1.48–1.94) 1.58 (1.34–1.86) n = 48 940 with information about prenatal and age-seven exposure; n = 49 014 with information about
prenatal exposure; n = 51 078 with information about age-seven exposure
a Reference category b Adjusted for sex of child, mother’s age at birth, mother’s and father’s history of psychiatric, cognitive or
behavioral problems as a child, combined socio-occupational status, gestational age, mother’s prenatal stress,
and child breastfed up to 6 months of age
c OR for prenatal exposure adjusted for age-seven exposure.
d OR for age-seven exposure adjusted for prenatal exposure.
Page 18 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
19
Table 3. Associations of mother’s prenatal cell phone use characteristics with overall
behavioral problems at age 11 years in children who had prenatal exposure and were at
risk of behavioral problems at age 7 years (n = 20 206)
N (%) Unadjusted OR Adjusted ORa
Times spoken per day
0–1b 10 189 (50.4) 1.00 1.00
2–3 5655 (28.0) 0.94 (0.78–1.13) 0.84 (0.68–1.05)
4+ 2192 (10.9) 1.22 (0.95–1.56) 1.11 (0.82–1.52)
Missing 2170 (10.7)
P for trendc 0.11 0.06
Percentage of time turned on
0b 1349 (6.7) 1.00 1.00
<50 2342 (11.6) 1.17 (0.80–1.71) 1.24 (0.79–1.95)
50–99 6110 (30.2) 1.11 (0.79–1.56) 1.03 (0.68–1.56)
100 10 258 (50.8) 1.32 (0.95–1.82) 1.14 (0.77–1.71)
Missing 147 (0.7)
P for trendc 0.49 0.39
Use of hands-free device
Nob 15 948 (78.9) 1.00 1.00
Rarely 2439 (12.1) 0.86 (0.67–1.11) 0.82 (0.60–1.11)
Often 1729 (8.6) 1.22 (0.95–1.56) 1.43 (1.06–1.92)
Missing 90 (0.5)
P for trendc 0.07 0.02
a Adjusted for sex of child, mother’s age at birth, mother’s and father’s history of psychiatric, cognitive or
behavioral problems as a child, combined socio-occupational status, gestational age, mother’s prenatal stress, child
breastfed up to 6 months of age, and age-seven exposure to cell phones. b Reference category
c Test for deviation from linear trend
Page 19 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
20
Table 4. Sensitivity analysis of associations between prenatal and age-seven cell phone exposure and overall
behavioral problems at age 11 years
Prenatal exposure only Age-seven exposure only
Both prenatal and age-
seven exposure
Cohort at riska
Unadjusted OR 1.40 (1.22–1.61) 1.39 (1.19–1.61) 1.69 (1.48–1.94)
Adjusted ORd 1.41 (1.20–1.66) 1.36 (1.14–1.63) 1.58 (1.34–1.86)
Cohort at risk excluding early adoptersb
Unadjusted OR 1.42 (1.23–1.64) 1.46 (1.24–1.72) 1.73 (1.50–2.00)
Adjusted ORd 1.39 (1.17–1.64) 1.38 (1.14–1.67) 1.55 (1.31–1.83)
All childrenc
Unadjusted OR 1.48 (1.33–1.65) 1.33 (1.1–1.51) 1.85 (1.66–2.06)
Adjusted ORd 1.44 (1.26–1.64) 1.27 (1.09–1.47) 1.61 (1.42–1.84)
Reference category is no exposure. a All children with exposure information excluding those with behavioral problem at age 7 years: n = 48 940 b All children with exposure information excluding those with behavioral problems at age 7 years and children born prior to 1999: n = 42 776 c All children with exposure information: n = 52 327 d Adjusted for sex of child, mother’s age at birth, mother’s and father’s history of psychiatric, cognitive or behavioral problems as a child, combined
socio-occupational status, gestational age, mother’s prenatal stress, child breastfed up to 6 months of age
Page 20 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Review Copy
21
Table 5. Comparison of analysis using cross-sectional data from Divan et al., 2010 with analysis
using current prospective data
Prenatal exposure only Age-seven exposure only
Both prenatal and age-
seven exposure
Cross-sectional analysisa
Unadjusted OR 1.5 (1.3–1.7)d 1.2 (1.1–1.4)
d 2.0 (1.7–2.1)
d
Adjusted ORc 1.4 (1.2–1.5) 1.2 (1.0–1.3) 1.5 (1.4–1.7)
Prospective analysisb
Unadjusted OR 1.4 (1.2–1.6) 1.4 (1.2–1.6) 1.7 (1.5–1.9)
Adjusted ORc 1.4 (1.2–1.7) 1.4 (1.1–1.6) 1.6 (1.3–1.9)
Reference category is no exposure. a Results from Divan et al., 2012: n = 41 541; “Cross-sectional analysis” of association between cell exposure assessed at age 7
years and behavioral problems assessed at age 7 years
b All children with exposure information excluding those with behavioral problem at age 7 years: n=48 940; “Prospective
analysis” using cell exposure assessed at age 7 years and behavioral problems assessed at age 11 years
c Adjusted for sex of child, mother’s age at birth, mother’s and father’s history of psychiatric, cognitive or behavioral problems as
a child, combined socio-occupational status, gestational age, mother’s prenatal stress, child breastfed up to 6 months of age
d Confidence were intervals were re-calculated for this analysis because they were not given in the original publication by Divan
et al., 2012.
Page 21 of 21
The American Academy of Pediatrics, 141 Northwest Point Blvd., Elk Grove Village, IL 60007
Confidential - Not for Circulation
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
top related